Crystallization, followed by subsequent structure determination, is a major step in understanding the structure- function relationship of macromolecules. Understanding macromolecule structure has become a key part in the development of new pharmaceuticals, and is a major area of NIH research. Crystallization however is also the rate limiting step, despite technological efforts to automate the set-up and crystallization data acquisition processes. The Phase I effort successfully demonstrated that a low cost epifluorescence type crystallization plate imaging system could be assembled, and that lead crystallization conditions could be obtained from apparently failed outcomes by intensity analysis of the fluorescence images. The method is based upon trace covalent labeling, defined as < 0.5% of the molecules being labeled, of the protein using a fluorescent probe that excites and emits in the visible spectrum. The major objectives of this proposal are to expand upon the Phase I results. First is the improvement of the software for the rapid scoring of the crystallization screen images, to expand that capability to include scoring for different crystallization outcomes (needle, plate, 3D crystal), and to further improve the scoring success rate. Second is the implementation of a multicolor fluorescence capability to make the instrument more suitable for the crystallization of macromolecule complexes, followed by the testing and demonstration of that capability. Third is to develop a labeling methodology suitable for use with integral membrane proteins. Fourth is to further refine the instrument and methods by continued use and testing in our laboratory. Experimentally, trace fluorescently labeled protein will be subjected to crystallization screens and the outcomes periodically imaged. Intensity-based image analysis, using the evolving software as it is developed during the proposal period, will be carried out. Precipitated conditions which show suitable scores based on the image analysis will be subjected to optimization screening, and based upon the Phase I effort results we expect a correlation between the scores obtained and subsequent crystallization. Previous research has shown that fluorescence can be a powerful aid in finding and identifying crystals in screening plates (Judge et al., 2005; Forsythe et al., 2006; Groves et al., 2007; Pusey et al., 2008). Crystallization gives the most densely packed state for a protein, and therefore trace fluorescently labeled protein will have the greatest fluorescence intensity relative to clear or precipitated outcomes. The covalently bound probe serves as a reporter to the protein's response to the solution conditions. Some precipitates showed 'bright spots' of fluorescence, and many of these outcomes were subsequently be optimized to crystallization conditions (Pusey et al., 2008; Phase I results). Thus intensity-based scoring of precipitation outcomes may be used to discriminate between non-productive and potentially productive precipitation results. Fluorescence intensity-based crystallization screen scoring is found to be fast, with image processing times currently 3 seconds.